2026-10-14T14:24:15Z https://www.scidb.cn/oai

10.57760/sciencedb.j00186.00252 2024-08-06T09:56:17Z

2024-08-06 Neural network model nuclide recognition training data set doi:10.57760/sciencedb.j00186.00252 en Using Geant4 to simulate NaI detectors, energy spectrum simulation experiments were conducted for 241Am, 192Ir, 226Ra, 133Ba, 60Co, 57Co, 137Cs, and 152Eu nuclides, obtaining initial 80 energy spectrum data (ten spectra for each nuclide) with 1024 channels (0-3MeV) for each spectrum. Considering the activity size of radioactive nuclides and the influence of environmental noise in actual measurements, the nuclide energy spectrum dataset is expanded by adding random noise and amplitude transformation. At the same time, the simulated nuclide energy spectrum is randomly combined to obtain a mixed energy spectrum of multiple nuclides. The dataset is divided into a validation set(validation) and a training set (database)  Gamma spectrum data; training set; validation set Fei Li Jiang Jianfeng PUBLIC https://creativecommons.org/licenses/by/4.0/ dataset http://www.doi.org/10.1007/s41365-024-01564-5 Science Data Bank